TNF-α AND IL-10 GENE POLYMORPHISMS VERSUS CARDIOIMMUNOLOGICAL RESPONSES IN SUDDEN INFANT DEATH

Fetal and Pediatric Pathology, 27:149–165, 2008Copyright C© Informa Healthcare USA, Inc.ISSN: 1551-3815 print / 1551-3823 onlineDOI: 10.1080/15513810802077651

TNF-α AND IL-10 GENE POLYMORPHISMS VERSUS

CARDIOIMMUNOLOGICAL RESPONSES IN SUDDEN INFANT DEATH

Nasrin Perskvist, Karin Skoglund, and Erik Edston � National Board of ForensicMedicine, Department of Forensic Medicine, Linkoping Division, Linkoping, Sweden

Gerd Backstrom, Irene Lodestad, and Ulla Palm � National Board of ForensicMedicine, Department of Forensic Genetics, Linkoping Division, Linkoping, Sweden

� We hypothesized that genetic variations of cytokines could contribute to the risk of developinga fatal immunological reaction in the heart of infants. Thus, tumor necrosis factor (TNF)-αand interleukin (IL)-10 gene polymorphisms versus induction of cardioimmunologxical responsesin victims of sudden infant death syndrome (SIDS) were explored. We genotyped 35 infants (23cases of SIDS and 12 infants with a known cause of death), and 100 healthy adult controls forIL-10 –1082 G/A, -592C/A and TNF-α -238 G/A, -308 G/A. We found a higher frequency ofthe ATA haplotype and ATA/ATA genotype of IL-10 associated with SIDS (13%). The frequency ofhomozygote infants for IL-10 haplotypes in SIDS was higher (52%) than the control group (34%).All SIDS cases were homozygotice for the TNF-α -238 G allele and 20 infants were homozygous forthe TNF-α -308 G allele in the same group. None of the infants with higher levels of infiltrated T-cells(n = 8) was homozygous for IL-10 gene polymorphisms, whereas in contrast 3 cases of the 6 thatdisplayed higher levels of cardiac mast cells were homozygous. In this study, the increased numberof interstitial T-cells, mast cells, and macrophages in the myocardial interstitium demonstrated nocorrelation with the genotype for either cytokines.

Keywords cytokines, gene polymorphism, myocardium, immunological competentcells, SIDS

INTRODUCTION

Sudden infant death syndrome (SIDS) still accounts for considerablenumbers of unexpected infant deaths in many countries. While numeroustheories have been advanced to explain these events, it is increasingly clearthat SIDS results from the complex interaction of a variety of heritablefactors interacting with exogenous factors [1]. SIDS probably has a complex

This study was supported by grants from the National Board of Forensic Medicine and the Stinaand Birger Johansson’s Foundation (L06-90346). The Local Board of Medical Ethics approved the study.

Address correspondence to Nasrin Perskvist, National Board of Forensic Medicine, Department ofForensic Medicine, Linkoping Division, Sweden Artillerigatan 12, SE 581 33, Linkoping, Sweden. E-mail:[email protected]

149

Feta

l Ped

iatr

Pat

hol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Cal

ifor

nia

Irvi

ne o

n 10

/31/

14Fo

r pe

rson

al u

se o

nly.

150 N. Perskvist et al.

etiology and is not a single disease, so that different genetic influences[2–4] might be important in different causes of death. All genes-encodingproteins involved in inflammatory responses are candidates for determiningthe human genetic background responsible for interindividual differencesin systemic inflammation and allergy. Infection, inflammation, and allergicresponses might be involved in the mechanisms underlying SIDS [5, 6].However, the problem is that death occurs suddenly and insufficientpathological findings at the postmortem examination make it difficult toexplain the cause of death. Cytokine polymorphism can alter the courseof diseases through alterations of the innate immune pathway. In additionto interleukin-10 (IL-10) and tumor necrosis factor-alpha (TNF-α), othercytokines also have been implicated in the development of pathologicalresponses in infants [7]. This study only attempts to cover the area relatingto TNF-α and IL-10.

IL-10 potentially inhibits inflammatory immune reactions in humans,primarily by inhibiting the immune activation of myocytes, dendritic cells,macrophages and their secretion of inflammatory mediators, thereby limit-ing their ability to act as antigen-presenting cells (APC) and to promote theactivation of Th1 type inflammatory T-cells [8]. The cells of hemopoieticlineages have the ability to generate TNF-α. Sensitized mast cells are knownto store a number of cytokines including TNF-α within their granules andto release them upon antigen presentation, thus making these a pivotal celltype, among others, in severe allergic asthmatic responses to allergens [9].TNF-α is known to induce fever that influences respiration in infants [10], aswell as somnolence and hypoxia, symptoms known to be related to suddeninfant deaths [7].

Myocardial TNF is an autocrine contributor to myocardial dysfunctionthrough direct disruption of contractility and induction of myocytic apopto-sis. The mechanism of TNF-induced contractile dysfunction appears to oc-cur through depression of systolic function by disrupting calcium-inducedcalcium release of the sarcoplasmic reticulum. In fact, TNF disrupts L-typechannel-induced calcium influx and thereby depresses calcium transients[11]. The resident myocardial macrophages and mast cells, and even car-diomyocytes themselves, synthesize TNF [11, 12]. One of the functions ofthe anti-inflammatory cytokine IL-10 is suppression of proinflammatory cy-tokine TNF-α activity.

There are many precedents for the particular importance of the TNF-α and IL-10 genotypes in the course of various diseases and the differentpolymorphisms in the genes of these cytokines have been associated withboth prevalence and severity of disease. The IL-10 gene is located on chro-mosome 1q31-32 [2], and TNF-α has been mapped to the class III regionof the MHC between the HLA-B and DR loci on chromosome six [13]. Inthe promoter region of both cytokines several single nucleotide polymor-phisms (SNPs) have been reported [14], and even though several of them

Feta

l Ped

iatr

Pat

hol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Cal

ifor

nia

Irvi

ne o

n 10

/31/

14Fo

r pe

rson

al u

se o

nly.

Affect of Gene Polymorphisms on Cardiac Functions 151

are in linkage disequilibrium, it is reasonable to assume that a number areof biological significance.

The most investigated SNPs in IL-10 are situated at position –1082 (A/G),-819 (C/T), and –592 (C/A) upstream from the transcriptional start site [14].IL-10 promoter polymorphisms among humans affect levels of IL-10 secre-tion and susceptibility to autoimmune diseases and related immune condi-tions [15, 16]. In particular, three linked SNPs located within 1100 bp ofthe transcription site characterize three haplotypes that we shall refer tohere as ATA, ACC, and GCC defined by G or A at –1082, C or T at –819,and C or A at –592 (numbering is relative to the transcription start site).Consistent with this idea, the ATA, ACC, and GCC haplotypes are associatedwith low, medium, and high levels of IL-10 secretion, respectively [17, 18].In contrast to the higher IL-10 expression in ACC and GCC haplotypes, thelower IL-10 expressing ATA haplotype is epidemiologically associated withthe development of autoimmune diseases [19, 20]. The haplotype GTA isextremely rare.

Several polymorphism areas are documented in the promoter regionof the TNF-α gene. In particular, two common biallelic variants at the –238(G or A) and –308 (G or A) positions of the TNF-α promoter region havebeen the first to receive attention [21, 22]. The −308A SNP is reported tohave an increased frequency in autoimmune and inflammatory diseases andis associated with stronger transcriptional activation than the G allele [23].Two polymorphisms, in particular at positions –308 and –238, are reportedlycapable of altering TNF-α expression [23].

Therefore, we hypothesize that in some situations, an infant with an un-favorable IL-10 and/or TNF-α genotype exhibits aberrant IL-10 and/or TNF-α production. This, in turn, leads to an imbalance in the immune responseand renders the infant unable to cope with infection. Thereby, the purposeof our study was to elucidate whether genetic polymorphisms at positions–1082 and –592 in the promoter region of IL-10 and –308 and -238 in thepromoter region of TNF-α were associated with the induction of cardioim-munological responses. Sections from the hearts of infants were examinedto determine the number of interstitial T-cells, mast cells, and macrophagesas well as the expression of myocardial TNF-α, and those compared with tothe haplotype and the genotype of the infants.

MATERIALS AND METHODS

Study Subjects

This study included material from 35 infants between 4 and 476 daysold. The infants had been subjected to detailed postmortem investigationat the Department of Forensic Medicine in Linkoping, Sweden. The causeof death was diagnosed as SIDS in 23 cases. The remaining 12 infants had

Feta

l Ped

iatr

Pat

hol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Cal

ifor

nia

Irvi

ne o

n 10

/31/

14Fo

r pe

rson

al u

se o

nly.


died from various diseases and unnatural causes, comprising 7 accidentaldeaths, 1 premature, 3 poisonings, and 1 case of epiglottitis. The DNAdata for comparison came from samples from 100 healthy living controlsfrom the same ethnic background that had been analyzed and documentedin the Department of Forensic Genetics in Linkoping, Sweden. Ethicalconsent was obtained for analysis of blood and paraffin-embedded tissues.All infants and controls were from the Swedish Caucasian ethnic subgroup.

Collection of Samples

A thorough forensic investigation of the infants was performed in allcases and included microscopic evaluation of internal organs, bacteriologi-cal culture from blood, lungs, and spinal fluid and in some cases from thenasopharynx, and toxicological analysis for alcohol and drugs (performedat the National Forensic Chemistry Laboratory, Linkoping, Sweden). Bloodwas sampled, mainly from the heart chambers and congested great vessels.After centrifugation, serum and blood were immediately frozen at –20◦C un-til analysis. Due to the limited amount of blood, the complete set of analysescould not be performed in all 35 infants.

DNA Extraction

DNA was extracted from blood using the commercially available QIAmpDNA Mini Kit (Qiagen), according to the manufacturer’s protocol. In pa-tients where blood was not available, DNA was extracted from formalin-fixedparaffin-embedded liver or heart sections. For each case, 1 or 2 0.02 mmslices were excised and placed into microfuge tubes. The tissues were lysedin the presence of proteinase K in ATL buffer at 56◦C for 2 hr, and DNA wasextracted using QIAmp DNA Mini Kit. DNA was eluted in 100 µl distilledwater (Milli-Q, Millipore). The concentration of DNA was measured byreal-time PCR using a QuantifilerTM Human DNA Quantification Kit(Applied Biosystems) and BioRad IQTM5 Real-Time PCR Detection System.

PCR Amplification and Sequencing

Fragments containing the promoter polymorphisms of IL-10 and TNF-αwere amplified by PCR. DNA from tissues was amplified in two short frag-ments containing IL-10 –1082 G/C and -592 C/A, respectively, and anothershort fragments containingTNF-α –308 G/A and –238 G/A polymorphisms.DNA from blood was amplified in two longer fragments containing IL-10–1082, -819, -592 and TNF-α –308, -238. Each sample was included in twoPCR reactions with different sequence specific primers. Two types of primerswere utilized, the long fragments in blood samples and the short fragmentsin the amplification of tissue-derived DNA (Table 1). Amplifications were

Feta

l Ped

iatr

Pat

hol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Cal

ifor

nia

Irvi

ne o

n 10

/31/

14Fo

r pe

rson

al u

se o

nly.

TA

BL

E1

Prim

erSe

quen

ces

Are

a/se

quen

ces

Prim

erse

quen

ces

Prod

uctl

engt

h(b

p)

Shor

tfra

gmen

tcon

tain

ing

Forw

ard

prim

er5′

-AC

AC

AC

AC

AC

AA

AT

CC

AA

GA

CA

A86

IL-1

0–1

082

Rev

erse

prim

er5′

-GG

AG

GT

CC

CT

TA

CT

TT

CC

TC

TT

ASh

ortf

ragm

entc

onta

inin

gFo

rwar

dpr

imer

5′-G

CC

TG

GA

AC

AC

AT

CC

TG

TG

AC

96IL

-10

–592

Rev

erse

prim

er5′

-CT

AA

AT

AT

CC

TC

AA

AG

TT

CC

CA

AG

CA

Prim

ers

pair

sof

TN

F-α

Forw

ard

prim

er5′

-CC

AA

AA

GA

AA

TG

GA

GG

CA

AT

AG

155

shor

tfra

gmen

tR

ever

sepr

imer

5′-C

CC

TC

AC

AC

TC

CC

CA

TC

CT

Prim

erpa

irs

ofIL

-10

Forw

ard

prim

er5′

-CA

CA

AA

TC

CA

AG

AC

AA

CA

CT

AC

TA

AG

594

lon

gfr

agm

ent

Rev

erse

prim

er5′

-CT

AA

AT

AT

CC

TC

AA

AG

TT

CC

CA

AG

CA

Prim

erpa

irs

ofT

NF-

αFo

rwar

dpr

imer

5′-T

CT

GA

AG

CC

CC

TC

CC

AG

TT

CT

AG

TT

CT

239

lon

gfr

agm

ent

Rev

erse

Prim

er5′

-CA

CA

CA

AG

CA

TC

AA

GG

AT

AC

CC

CT

CA

Sequ

ence

prim

erPr

imer

sequ

ence

sIL

-10

–108

2G

/AFo

rwar

dpr

imer

5′-C

AA

CA

CT

AC

TA

AG

GC

TT

CT

TT

GG

GA

–IL

-10

–819

C/T

Forw

ard

prim

er5′

-GT

GT

AC

CC

TT

GT

AC

AG

GT

GA

TG

TA

A–

IL-1

0–5

92C

/AFo

rwar

dpr

imer

5′-A

TC

CT

GT

GA

CC

CC

GC

CT

GT

–T

NF-

a–3

08G

/AFo

rwar

dpr

imer

5′-A

AA

AA

AA

AA

AA

GA

GG

CA

AT

AG

GT

TT

TG

AG

GG

GC

AT

G–

TN

F-a-

238

C/T

Rev

erse

Prim

er5′

AA

AA

AA

AA

AA

AA

TC

AC

AC

TC

CC

CA

TC

CT

CC

CT

GC

TC

–

153

Feta

l Ped

iatr

Pat

hol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Cal

ifor

nia

Irvi

ne o

n 10

/31/

14Fo

r pe

rson

al u

se o

nly.


performed in 25 µL reactions with AmpliTaq©RDNA Polymerase (AppliedBiosystems) with a final concentration of 0.02 U AmpliTaq, 12 mM Tris-HCl pH 8.3, 60 mM KCl, 1.8 mM MgCl2, 200 µM dNTPs, and 0.08 µM ofeach primer. The cycling conditions were 94◦C for 2 min denaturation fol-lowed by 32 cycles at 94◦C for 30 sec, at 55.5◦C for 50 sec, and at 72◦ for30 sec. For tissue, DNA amplification was repeated in 32 cycles with 10 µLPCR product as target DNA. Excess dNTPs and primers were removed byadding 2.5 U SAP (shrimp alkaline phosphatase) and 1 U Exo1 (GE Health-care) to 7.5 µL PCR product and incubated at 37◦C for 1 hr followed by 15min at 75◦C. The SNPs were detected by single base extension using ABIPrism©RSNaPshotTMMultiplex Kit (Applied Biosystems).

Polymorphisms of IL-10 and TNF-α were analyzed separately for DNAfrom tissue. DNA from blood was analyzed in one reaction. The reactionmixture contained 2 µL SNaPshot multiplex reaction mix, 3-4 µL PCRproducts, 1 µL pooled sequence primers (0.015–0.022 µM), and waterto a final volume of 10 µL. The reactions were run in a GeneAmp 9700for 25 cycles at 96◦C for 10 sec, at 50◦ for 5 sec, and at 60◦ for 30 sec.Unincorporated fluorescence-labeled ddNTPs were removed by treatmentwith 1U SAP at 37◦C for 1 hr and 15 min at 75◦C. Then 2 µL SNaPshotproduct were mixed with 10 µL HI-DI formamide (Applied Biosystems)and analyzed by capillary electrophoreses using ABI Risma©R 310 withPOP4-polymer (Applied Biosystems).

Immunohistochemical Analysis

Immunohistochemical staining of CD4+, CD8+ T-cells, intracellularlocalization of tryptase (a marker of mast cells), and expression of TNF-αwere performed using monoclonal antibodies. The methods appliedfor immunohistochemical staining and the antibodies used have beendescribed in detail elsewhere [12, 24]. For visualization of all antibodies, aperoxidase-labelled secondary antibody was used; diaminobenzidine (DAB)was applied as chromogen. The sections were then counterstained withlight green. Antibody to tryptase (NOVO Castra Laboratories) was used toidentify cardiac mast cells. The anti-CD4+, and -CD8+ T-cells (Dako) wereused to distinguish the T-cell population. The anti-CD68 (Dako) and anti-TNF-α (ab-cam) were used to localize the macrophages and TNF-positivecells in myocardium, respectively. Human tonsil tissue was used as a positivecontrol and omitted antibodies as a negative control.

For each case, 4 sequential sections from the left and right ventricularwall and septum, along with a transverse section of the whole heart, werestained, the cells were counted in 20 randomized high power fields (HPF,magnification ×400), and the mean value was calculated. TNF-positivecells were estimated in the same manner. Two independent observersperformed these microscopic examinations. To verify the results from

Feta

l Ped

iatr

Pat

hol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Cal

ifor

nia

Irvi

ne o

n 10

/31/

14Fo

r pe

rson

al u

se o

nly.


immunohistochemical labelling of mast cells, the tissues were simultane-ously stained with Toluidine blue for counting mast cells.

Conventional Histological Staining

Microscopic sections including all the cardiac sections and sections fromthe brain, lungs, liver, kidneys, proximal coronary arteries from the threemain branches, and laryngeal wall were stained with hematoxylin-erythrosin-saffron and Mallory’s PTAH.

Statistic Analysis

The data were analyzed statistically using SPSS version 11.5 (Chicago,IL, USA). The data obtained from SIDS, the victims of unnatural death, andthe controls were compared using the Mantel-Haenszel χ2 test and Fisherexact test (when appropriate). The data were tested for their fit to Hardy-Weinberg equilibrium by χ2test. Spearman’s rank correlation test was usedto examine correlation between the number of infiltrated and/or residentsof cardiac mast cells, T cells, macrophages, and the type of cytokine genepolymorphisms. Because several comparisons were performed between thegroups, the cut-off for significant p value was set to 0.025.

RESULTS

Fully 23 cases of SIDS (10 females and 13 males) and 12 cases ofinfants (4 females and 8 males) with a defined cause of death (DCD) werecompared with 100 cases of living healthy individuals. The mean age ofSIDS cases was 220 days. Using the modified DNA extraction and PCRconditions described above and SnaPShot technique for final sequencing,it was possible to successfully and reliably genotype individuals usingparaffin-embedded tissue samples. In cases in which blood from the sameindividual was available, we performed analysis of the blood also to increasethe validity of the results.

Pathological, Histological and Microbiological Findings

In the SIDS group (n = 23), the lungs were congested with edema andhemorrhages in all cases. Petechial hemorrhages in the pleura, pericardium,and thymic capsule were seen in less than 50% of SIDS cases. Organweights were all within the normal range [25]. Bacterial cultures from cere-brospinal fluid, blood, and both lungs either showed no growth of bacteriaor fungi or in a minority of cases moderate growth of diverse nonpathogenicbacteria. These were regarded as contaminants. Furthermore, histological

Feta

l Ped

iatr

Pat

hol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Cal

ifor

nia

Irvi

ne o

n 10

/31/

14Fo

r pe

rson

al u

se o

nly.


TABLE 2 Distribution of IL-10 Haplotypes of IL-10 in SIDS

Haplotype SIDS (n = 23) DCD (n = 12) HC (n = 100)

GCC 23 (50%) 10 (42%) 102 (51%)ACC 9 (20%) 9 (37%) 51 (25.5%)ATA 14 (30%) 5 (21%) 47 (23.5%)

DCD = defined causes of death; HC = healthy controls.

examination of various tissues from these cases did not show any corre-sponding inflammatory reaction in the lungs, brain, or elsewhere.

Distribution of IL-10 Promoter Gene Polymorphism

We found a higher frequency of the ATA haplotype and ATA/ATA geno-type of IL-10 associated with SIDS cases (13%, n = 23), as opposed to 2%(n = 100) in healthy living control cases (Tables 2 and 3). However, we werenot able to find any significant differences (χ2 Mantel-Haenszel and/orFisher exact test) between SIDS and DCD compared with the healthy con-trols (HC) population, in spite of the results of several other investigationsthat support the hypothesis that the ATA haplotype of IL-10 was associatedboth with SIDS and sudden unexpected death due to infections [13, 26,27]. Three haplotypes of IL-10 polymorphisms exist, combining the SNPs–1082, -819, and –592, namely ATA, ACC, and GCC, respectively. A survey ofthe SNP haplotype of IL-10 of all three groups is presented in Table 2. Whenthe haplotype frequencies between the controls and the infants (SIDS andDCD) were analyzed, we could not find significant differences (Table 2).Among the homozygous infants in the SIDS group, there were 7 cases withGCC, 3 with ATA and 2 with ACC homozygous genotypes (see Table 5).Of the 23 SIDS cases 12 were homozygous (52%)(see Table 5), whereas34 of 100 cases (34%) of the control group were homozygous for IL-10haplotypes. Analysis by the χ2 test resulted in no significant differencesbetween the SIDS group and the control group in respect to AA genotypeof IL-10 –592. The differences within the -1082 G/A genotype in DCD werenot statistically applicable due to the small number of cases (n = 12).

Distribution of TNF-α Promoter Genotype Polymorphism

Table 4 shows the frequencies of genotypes and alleles for –238 Aor Gand –308 A or G TNF-α gene in SIDS, DCD, and the healthy living controls.According to the results of this study, none of the infants carried the Aallele, AA, or GA genotype of TNF-α -238 whereas a few individuals in thecontrol group carried these genotypes (Table 4). All 23 infants in the SIDSgroup were homozygous for the TNF-α -238 G allele as well as 20 casesfor the TNF-α -308Gallele. Correspondingly 90% of the SIDS cases were

Feta

l Ped

iatr

Pat

hol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Cal

ifor

nia

Irvi

ne o

n 10

/31/

14Fo

r pe

rson

al u

se o

nly.

TA

BL

E3

Obs

erve

dFr

eque

nci

esfo

rA

llele

san

dG

enot

ypes

ofIL

-10

inSI

DS

Gen

otyp

e–1

082

Alle

le-1

082

Gen

otyp

e–5

92A

llele

-592

Popu

lati

ons

(n)

G/G

G/A

A/A

GA

C/C

C/A

A/A

CA

SID

S(n

=23

)7(

30%

)9(

40%

)7(

30%

)23

(50%

)23

(50%

)12

(52%

)8(

35%

)3(

13%

)32

(70%

)14

(30%

)D

CD

(n=

12)

1(8%

)8(

67%

)3(

25%

)10

(42%

)14

(58%

)7(

58%

)5(

42%

)0

19(7

9%)

5(21

%)

HC

(n=

100)

25(2

5%)

52(5

2%)

23(2

3%)

102(

51%

)98

(49%

)55

(55%

)43

(43%

)2(

2%)

153(

76.5

%)

47(2

3.5%

)

DC

D=

defi

ned

caus

esof

deat

h;H

C=

hea

lth

yco

ntr

ols.

157

Feta

l Ped

iatr

Pat

hol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Cal

ifor

nia

Irvi

ne o

n 10

/31/

14Fo

r pe

rson

al u

se o

nly.

TA

BL

E4

TN

F-α

Prom

oter

Gen

otyp

ean

dA

llele

Freq

uen

cies

amon

gSI

DS

Gen

otyp

e–2

38A

llele

-238

Gen

otyp

e–3

08A

llele

-308

Popu

lati

ons

(n)

G/G

G/A

A/A

GA

G/G

G/A

A/A

GA

SID

S(n

=23

)23

(100

%)

00

46(1

00%

)0

18(7

8%)

5(2

2%)

041

(89%

)5

(11%

)D

CD

(n=

12)

12(1

00%

)0

024

(100

%)

07

(58%

)5

(42%

)0

19(7

9%)

5(2

1%)

HC

(n=

100)

97(9

7%)

3(3

%)

019

7(9

8.5%

)3

(1.5

%)

63(6

3%)

34(3

4%)

3(3

%)

160(

80%

)40

(20%

)

DC

D=

defi

ned

caus

esof

deat

h;H

C=

hea

lth

yco

ntr

ols.

158

Feta

l Ped

iatr

Pat

hol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Cal

ifor

nia

Irvi

ne o

n 10

/31/

14Fo

r pe

rson

al u

se o

nly.


TABLE 5 Distribution of TNF-α and IL-10 Gene Polymorphisms Versus CardioimmunologicalResponses

T-cells TNF-α genotypes IL-10 genotypes IL-10

Cases CD4+ CD8+ Macrophages Mast cells −238 −308 −1038 −592 haplotypes

1-S 38 16 1 6 G GA GA CA GCC/ATA2-S 41 12 2 11 G G GA C GCC/ACC3-S 25 15 0 5 G G GA CA GCC/ATA4-S 18 15 0 3 G GA GA C GCC/ACC5-S 24 9 2 8 G G A CA ACC/ATA6-S 17 9 0 53 G G GA CA GCC/ATA7-S 9 8 16 41 G G A A ATA8-S 4 6 2 20 G G G C GCC9-S 4 2 2 20 G G GA C GCC/ACC10-S 13 4 1 15 G G A CA ACC/ATA11-S 15 5 1 14 G G G C GCC12-S 6 4 0 14 G G A C ACC13-S 4 2 4 14 G G G C GCC14-S 7 2 5 13 G GA A A ATA15-S 6 3 0 12 G G A C ACC16-S 3 2 1 11 G GA GA CA GCC/ATA17-S 16 11 3 4 G G GA CA GCCATA18-S 12 4 4 4 G G G C GCC19-S 8 11 1 7 G GA G C GCC20-S 13 1 6 6 G G G C GCC21-S 4 3 2 4 G G G C GCC22-S 6 3 1 3 G G GA CA GCC/ATA23-S 3 2 2 2 G G A A ATA24-D 30 43 21 3 G GA GA C GCC/ACC25-D 20 31 32 4 G GA A CA ACC/ATA26-D 21 17 0 8 G G GA C GCC/ACC27-D 10 14 29 23 G G A CA ACC/ATA28-D 10 12 2 20 G GA G C GCC29-D 6 6 17 10 G GA A C ACC30-D 5 4 9 6 G G GA C GCC/ACC31-D 6 2 0 15 G G GA C GCC/ACC32-D 1 4 12 19 G GA GA C GCC/ACC33-D 1 0 1 8 G G GA CA GCC/ATA34-D 0 0 1 5 G G GA CA GCC/ATA35-D 0 0 2 3 G G GA CA GCC/ATA

S = SIDS (n = 23), D = DCD (n = 12).Number of T-cells, macrophages, and mast cells is presented as the mean number of the positivecells/20 high power field.

homozygous for the G allele of –238 and –308 as compared with 80% of thecontrol subjects (Tables 4 and 5). No significant differences were detectedin the frequency of genotypes and alleles among SIDS, DCD, and controlcases when the data were analyzed by gene count and contingency tables(χ2 test). According to the χ2 test, the distribution of genotypes was inHardy-Weinberg equilibrium.

Feta

l Ped

iatr

Pat

hol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Cal

ifor

nia

Irvi

ne o

n 10

/31/

14Fo

r pe

rson

al u

se o

nly.


Immunohistochemical Analysis

The aim of this analysis was to examine the possible relationships be-tween the cytokine polymorphisms and the induction of in situ cardiac in-flammatory reactions. We pooled the SIDS cases with infants who died of un-natural death to gain a larger infant group (Table 5). The cardiac sectionsfrom all infants were immunoassayed for T-cells, mast cells, macrophages,and the expression of TNF-α. According to our results and the results ofother investigators, the mean standard value of myocardial T-cells (the sumof CD4+ and CD8+) was found to be 30 T-cells/20 high power field (HPF)(magnification ×400) [28, 29]; the value for myocardial macrophages wasfound to be 20–30 macrophages/20 HPF [30]; the mean for heart mast cellswas 16–18 mast cells/20 HPF [24]. A large number of cardiac mast cellsshowed positive staining for TNF-α expression, while none of the cardiomy-ocytes expressed this cytokine.

As a control for reliability of the method, we have applied the same tech-nique to investigate the myocytic production of TNF-α in the hearts of thevictims of drug overdose, and we found the expression of cardiomyocyticTNF-α [12]. Figure 1 shows the distribution of CD4+, CD8+, macrophages,and mast cells in cardiac sections of infants. Estimation of the mean num-ber of interstitial cells in cardiac sections revealed that 5 cases of SIDS and3 cases of DCD infants had higher numbers of T-cells, 4 cases of SIDS and2 cases of DCD individuals showed enhancement of cardiac mast cells, andonly 2 cases of DCD exhibited a higher number of macrophages, whencompared with the standard values, respectively (Table 5). Furthermore,the number of toluidine-stained mast cells was underestimated by approxi-mately 20% compared with the number of immunologically labelled cells.

Analysis Between Cytokine Gene Polymorphisms and Myocardial

Enhancement of CD4+, CD8+, Macrophages, and Mast Cells

To gain a better understanding of the immunological role of IL-10and TNF-α gene polymorphisms in the hearts of these infants, we charac-terized and measured the enhanced numbers of interstitial T-cell subsets,macrophages, and mast cells in the myocardium and compared with thegenotype and allele polymorphisms of TNF-α and IL-10. Several T-cell sub-sets have been previously shown to produce IL-10, including Th2 effectorcells and various regulatory and memory T-cell subsets [31, 32]. Addition-ally, IL-10 is one of the important immunoregulatory cytokines that mod-ulate the balance between Th1 and Th2 immunresponses. In fact, the dif-ferentiation and enhancement of CD4+ and/or CD8+ subsets of T-cells is aresponse to viral or bacterial proteins, respectively.

In our population of infants, we demonstrated (Table 5) that in-creased infiltration of CD4+, CD8+, mast cells, and macrophages into the

Feta

l Ped

iatr

Pat

hol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Cal

ifor

nia

Irvi

ne o

n 10

/31/

14Fo

r pe

rson

al u

se o

nly.


FIGURE 1 Representative photo micrographs of inflammatory cell infiltrates in the myocardium ofthe victims of sudden infant death. Myocardial sections were stained immunohistochemically using anti-bodies to CD4+ (a section from interventricular septum), CD8+ (a section from left ventricle posteriorwall), tryptase to detect mast cells (a section from posterior wall of right ventricle), and CD68, a markerof macrophages (a section from endocardium of right ventricle) (magnification × 100).

myocardium are IL-10 haplotype- and genotype-independent, occurringsimilarly in all three of the most common haplotypes of the polymorphicIL-10 promoter. None of the infants with a higher level of interstitial (orinfiltrating) T-cells (n = 8) was homozygous for IL-10 gene polymorphisms,in comparison to 3 cases of 6 with a higher level of cardiac mast cells thatwere homozygous. The accumulation of all cell types in the myocardialinterstitium was independent of TNF-α genotypes and allele (Table 5).

DISCUSSION

If the induction of increased levels of local IL-10 and/or TNF-α isassociated with induction of fatal cardioimmunopathological reactions,then it is a logical step to consider genotypic analysis of the deceased andtheir families to see if SIDS was more common in those who have the abilityto generate different levels of cytokines by virtue of polymorphic variants.One such candidate is the polymorphism of TNF-α. The –308 and –238TNF-α promoter polymorphisms are biallelic (G to A) and the A genotype is

Feta

l Ped

iatr

Pat

hol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Cal

ifor

nia

Irvi

ne o

n 10

/31/

14Fo

r pe

rson

al u

se o

nly.


associated with higher amounts of TNF-α on stimulation in-vivo and ex-vivo[33]. A number of independent reports have also indicated an associationof –308 TNF-α with asthma [34, 35]. TNF-α is a chemotactic cytokine andin the heart, myocardial mast cells, macrophages, and myocytes themselvesare known to synthesize and/or store TNF-α [11, 12]. Mast cells are themost fundamental cells in initiation of allergic and anaphylactic reactions[24]. Elevated in situ levels of TNF-α through infiltrating and/or residentcardiac mast cells and macrophages contribute to postischemic myocardialdysfunction via direct depression of contractility and induction of myocyticapoptosis. This cascade might have fatal consequences in infants.

In our population of SIDS cases, we observed a minor number of the-308 Agenotype and allele compared with G -308 SNP within groups; how-ever, there were no significant differences between healthy individuals andour SIDS cases. Compared with the reference value for cardiac mast cells,6 infants showed higher numbers of myocardial mast cells per square area.Yet, there was no association between the genotype or allele of TNF-α G/A-308 and -238 and the higher number of cardiac mast cells in these individ-uals. However, an investigation of other gene polymorphism areas of TNF-αon a larger SIDS population might lead to more significant findings.

IL-10 is a regulatory cytokine that is mainly produced by macrophages,T-cells, B-cells, and eosinophils. The role of IL-10 in vivo is to limit inflamma-tory responses. A study [36] of patients with acquired pneumonia reportedthat the –1082 GG genotype was associated with more severe disease than thegenotypes –1082 GA and –1082 AA. The study showed a significantly higherG allele frequency in patients who died from pneumonia, whereas anotherinvestigation revealed that this allele protects against Epstein- Barr virus in-fection [37]. The contradictory results of several investigations reported anassociation between SIDS and a SNP of the IL-10 –592 A allele [14, 26, 27,38]. When IL-10 production was measured in culture supernatants, the AAgenotype was associated with lower IL-10 production in both patients andhealthy controls [38]. Fully 13% of our SIDS cases carried the genotype AAin the –592 IL-10 SNP compared with 2% in healthy controls. To find astatistically significant difference between the SIDS group and the healthycontrols would have required a larger number of SIDS cases.

IL-10 is a cytokine of T-cells, both in differentiation and proliferation.T-cell-derived IL-10 and T-cell secretion of IL-10 in the circulation and heartplay important roles in both the progression and/or inhibition of asthmaticreactions and dysfunction of cardiac myocytes [14, 38, 39]. In spite of thesepathological effects of IL-10 and the significant cardiac accumulation of T-cells (CD4+ and CD8+) in 8 of our cases, we could not demonstrate anassociation between IL-10 polymorphisms in SNP –1082 and/or –592 andSIDS. Moreover, 3 of the DCD cases (accidental deaths) showed a significantamounts of cardiac T cells, an observation that requires further postmortemexamination in these cases.

Feta

l Ped

iatr

Pat

hol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Cal

ifor

nia

Irvi

ne o

n 10

/31/

14Fo

r pe

rson

al u

se o

nly.


SIDS occupies a difficult position between the legal and medical systems.It has been suggested that SIDS is a multifactorial disorder influenced by de-velopmental, environmental, and biological risk factors. All mutations giv-ing rise to metabolic disorders known to be associated with life-threateningevents are possible candidates for genes involved in cases of SIDS, either asa cause of death or as a predisposing factor. Mutation in genes encoding car-diac ion channels, e.g., in the Long QT syndrome [40], and mutation in theprocesses of glucose [41] and fatty acid metabolism [42] may all be involvedin sudden death of infants. However it is necessary to distinguish betweenlethal mutations leading to death and polymorphisms that are normal genevariants. The controversial findings [43, 44] suggesting that IL-10 polymor-phisms are frequently higher in SIDS cases may indicate that in a given sit-uation, an infant with a certain IL-10 genotype may exhibit aberrant IL-10production. However, this does not confirm the assumption that this gene,which is involved in immune responses, is a cause of death in these infants.The polymorphisms of the C4 gene [45, 46], HLA-DR [47], and the thermalprotein-encoding gene [48] have been investigated in a small populationof SIDS cases. A significant association of these genetic polymorphisms andthe occurrence of sudden infant death has not been confirmed. Two inves-tigations, an American study with 87 SIDS cases [49] and a Japanese studywith 27 SIDS cases [50], investigated the regions of the serotonin transporterprotein gene that modulates gene expression. Both revealed associations be-tween the long alleles of the serotonin transport gene and SIDS in differentpercentages of the populations investigated.

To our knowledge there are no previous publications on associationbetween TNF-α gene polymorphisms and SIDS. The investigation ofthe relationships between cytokine polymorphisms and the induction ofimmunological reactions of the heart is the first step toward understandingthe impact of TNF-α and IL-10 gene polymorphisms on the fatal cardiacresponses with regard to SIDS. The higher frequency of the ATA andthe frequency of homozygote infants for IL-10 haplotypes, as well as thefrequency of homozygote individuals for the G allele of TNF-α, showedno relationships to the cardioimmunological reactions. Our SIDS groupmay affect cardiac function by a mechanism not related to the number ofinflammatory cells in the myocardium. however, these results deserve to beconfirmed by a larger SIDS population.

REFERENCES

1. Bajanowski T, Vege A, Byard RW, et al. Sudden infant death syndrome (SIDS)—standardised inves-tigations and classification: recommendations. Forensic Sci Int 165:129–43, 2007.

2. Eskdale J, Kube D, Gallagher G. A second polymorphic dinucleotide repeat in the 5′flanking region

of the human IL10 gene. Immunogenetics 45:82–83, 1996.3. Summers AM, Summers CW, Drucker DB, Hajeer AH, Barson A, Hutchinson IV. Association of

IL-10 genotype with sudden infant death syndrome. Hum Immunol 61:1270–1273, 2000.

Feta

l Ped

iatr

Pat

hol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Cal

ifor

nia

Irvi

ne o

n 10

/31/

14Fo

r pe

rson

al u

se o

nly.


4. Hoffmann SC, Stanley EM, Cox ED, DiMercurio BS, Koziol DE, Harlan DM, Kirk AD, Blair PJ.Ethnicity greatly influences cytokine gene polymorphism distribution. Am J Transplant 2:560–567,2002.

5. Blackwell C. Infection, inflammation and SIDS. FEMS 42:1–2, 2004.6. Goldwater PM. SIDS pathogenesis: pathological findings indicate infection and inflammatory re-

sponses are involved. FEMS 42:11–20, 2004.7. Vege A, Rognum TO. Sudden infant death syndrome, infection and inflammatory responses. FEMS

42:3–10, 2004.8. Lee JJ, Kim DH, Lee NY, Sohn SK, Kim JG, Kim HJ, Do YR, Park YH. Interleukin-10 gene poly-

morphism influences the prognosis of T-cell non-Hodgkin lymphomas. Br J Haematol 137:329–336,2007.

9. Anticevich SZ, Hughes JM, Black JL, Armour CL. Induction of human airway hyperresponsivenessby tumour necrosis factor-alpha. Eur J Pharmacol 284:221–225, 1995.

10. Riesenfeld T, Hammarlund K, Norsted T, Sedin G. Irregular breathing in young lambs and newborninfants during heat stress. Acta Pediatr 85:467–470, 1996.

11. Meldrum DR. Tumor necrosis factor in the heart. Am J Physiol 274:577–595, 1998.12. Perskvist N, Soderberg C, van Hage M, Edston E. Pathogenic role of cardiac mast cell activa-

tion/degranulation, TNF-α, and cell death in acute drug-related fatalities. Vasc Health Risk Manag.3:1–10, 2007.

13. Niro GA, Fontana R, Gioffreda D, et al. Tumor necrosis factor gene polymorphisms and clearanceor progression of hepatitis B virus infection. Liver Inter 25:1175–1181, 2005.

14. Opdal SH. IL-10 gene polymorphisms in infectious disease and SIDS. FEMS 42:48–52, 2004.15. Lin MT, Storer B, Martin PJ, Tseng LH, Gooley T, Chen PJ, Hansen JA. Relation of an IL-10 promoter

polymorphism to graft-versus-host disease and survival after hematopoietic-cell transplantation. NEngl J Med 349:2201–2210, 2003.

16. Howell WM. Cytokine polymorphisms, cancer susceptibility, and prognosis. Ame Soc Histocompat Im-munogen Quarterly 1st Quarter: 10–15, 2005.

17. Eskdale J, Gallagher G, Verweij CL, Keijsers V, Westendorp RGJ, Huizinga TWJ. IL-10 secretion inrelation to human IL-10 locus haplotypes. Proc Natl Acad Sci 95:9465–9470, 1998.

18. Hoffman SC, Stanley EM, Cox ED, et al. Association of cytokine polymorphic inheritance and invitro cytokine production in anti-CD3/CD28-stimulated peripheral blood lymphocytes. Transplanta-tion 72:1444–1450, 2001.

19. Bidwell J, Keen L, Gallagher G, et al. Cytokine gene polymorphism in human disease: on-linedatabases. Genes Immun 1:3–19, 1999.

20. Haukim, N, Bidwell J, P. Smith AJ, et al. Cytokine gene polymorphism in human disease: on-linedatabases. Genes Immun 3:313–330, 2002.

21. D’Alfonso S, Richiardi PM. A polymorphic variation in a putative regulation box of the TNFA pro-moter region. Immunogenetics 39:150–154, 1994.

22. Wilson AG, di Giovine FS, Blakemore AIF, Duff GW. Single base polymorphism in the human tumornecrosis factor alpha (TNF-alpha) gene detectable by NcoI restriction PCR product. Hum Mol Genet1:353–357, 1992.

23. Domenico L, Annoni G, Licastro F, et al. Tumor necrosis factor-α−308A/G polymorphism is associ-ated with age at onset of Alzheimer’s disease. Mech Ageing Dev 127:567–571, 2006.

24. Perskvist N, Edston E. Differential accumulation of pulmonary and cardiac mast cell-subsets andeosinophils between fatal anaphylaxis and asthma death: a postmortem comparative study. ForensicSci Int. 169:43–49, 2007.

25. Thompson WS, Cohle SD. Fifteen-year retrospective study of infant organ weights and revision ofstandard weight tables. J Forensic Sci 49:575–585, 2004.

26. Moscovis SM, Gordon AE, Al Madani OM, et al. Interleukin-10 and sudden infant death syndrome.FEMS 42:130–138, 2004.

27. Korachi M, Pravica V, Barson AJ, Hutchinson IV, Drucker D. Interleukin 10 genotype as a risk fac-tor for sudden infant death syndrome: determination of IL-10 genotype from wax-embedded post-mortem samples. FEMS 42:125–129, 2004.

28. Dettmeyer R, Schlamann M, Madea B. Immunohistochemical techniques improve the diagnosis ofmyocarditis in cases of suspected sudden infant death syndrome (SIDS). Forensic Sci Int 105:83–94,1999.

Feta

l Ped

iatr

Pat

hol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Cal

ifor

nia

Irvi

ne o

n 10

/31/

14Fo

r pe

rson

al u

se o

nly.


29. Dettmeyer R, Baasner A, Schlamann M, et al. Role of virus-induced myocardial affections in suddeninfant death syndrome: a prospective postmortem study. Pediatr Res 55:947–952, 2004.

30. Azzawi M, Kan SW, Hillier V, Yonan N, Hutchinson IV, Hasleton PS. The distribution of cardiacmacrophages in myocardial ischaemia and cardiomyopathy. Histopathology 46:314–319, 2005.

31. Groux H, Cottrez F. The complex role of IL-10 in autoimmunity. J Autoimm 20:281–285, 2003.32. Yssel H, Malefyt RDW, Roncarolo MG, Abrams JS, Lahesmaa R, Spits H, deVries JE. IL-10 is produced

by subsets of human CD4+ T cell clones and peripheral blood T cells. J Immunol 149:2378–2384,1992.

33. Louis E, Franchimont D, Piron A, et al. Tumour necrosis factor (TNF) gene polymorphism influ-ences TNF-alpha production in lipopolysaccharide (LPS)-stimulated whole blood cell culture inhealthy humans. Clin Exp Immunol 113:401–406, 1998.

34. Thomas PS. Tumour necrosis factor-α: the role of this multifunctional cytokine in asthma. ImmunolCell Biol 79:132–140, 2001.

35. Moffatt MF, Cookson WO. Tumour necrosis factor haplotypes and asthma. Hum Mol Genet 6:551–554, 1997.

36. Schaaf BM, Boehmke F, Esnaashari H, Seitzer U, Kothe H, Maass M, Zabel P, Dalhoff K. Pneumo-coccal septic shock is associated with the interleukin-10-1082 gene promoter polymorphism. Am JRespir Crit Care Med 168:476–480, 2003.

37. Helminen M, Lahdenpohja N, Hurme M. Polymorphism of the interleukin-10 gene is associatedwith susceptibility to Epstein-Barr virus infection. J Infect Dis 180:496–499, 1999.

38. Stanilova SA, Miteva LD, Karakolev ZT, Stefanov CS. Interleukin-10-1082 promoter polymorphismin association with cytokine production and sepsis susceptibility. Intensive Care Med 32:260–266, 2006.

39. Hawrylowicz CM, O’Garra A. Potential role of interleukin-10-secreting regulatory T cells in allergyand asthma. Nat Rev Immunol 5:271–283, 2005.

40. Schwartz PJ, Stramba-Badiale M, Segantini A, et al. Prolongation of the QT interval and the suddeninfant death syndrome. N Engl J Med 338:1709–1714, 1998.

41. Burchell A, Lyall H, Busuttil A, Bell E, Hume R. Glucose metabolism and hypoglycaemia in SIDS. JClin Pathol 45:39–45, 1992.

42. Yokota I, Indo Y, Coates PM, Tanaka K. Molecular basis of medium chain acyl-coenzyme A dehy-drogenase deficiency: an A to G transition at position 985 that causes a lysine-304 to glutamatesubstitution in the mature protein is the single prevalent mutation. J Clin Invest. 86:1000–1003,1990.

43. Summers AM, Summers CW, Drucker DB, Hajeer AH, Barson A, Hutchinson IV. Association ofIL-10 genotype with sudden infant death syndrome. Hum Immunol 61:1270–1273, 2000.

44. Opdal SH, Opstad A, Vege A, Rognum TO. IL-10 gene polymorphisms are associated with infectiouscause of sudden infant death. Hum Immunol 64:1183–1189, 2003.

45. Schneider PM, Wendler C, Riepert T, et al. Possible association of sudden infant death with partialcomplement C4 deficiency revealed by post-mortem DNA typing of HLA class II and III genes. EurJ Pediatr 149:170–174, 1989.

46. Opdal SH, Vege A, Stave AK, Rognum TO. The complement component C4 in sudden infant death.Eur J Pediatr 158:210–212, 1999.

47. Keller E, Andreas A, Teifel-Greding J, et al. DNA analysis of HLA class II and III genes in suddeninfant death (SIDS). Beitr Gerichtl Med 48:285–290, 1990.

48. Rahim RA, Boyd PA, Ainslie Patrick WJ, Burdon RH. Human heat shock protein gene polymor-phisms and sudden infant death syndrome. Arch Dis Child 75:451–452, 1996.

49. Weese-Mayer DE, Berry-Kravis EM, Maher BS, Silvestri JM, Curran ME, Marazita ML. Sudden infantdeath syndrome: association with a promoter polymorphism of the serotonin transporter gene. AmJ Med Genet 117:268–274, 2003.

50. Narita N, Narita M, Takashima S, Nakayama M, Nagai T, Okado N. Serotonin transporter genevariation is a risk factor for sudden infant death syndrome in the Japanese population. Pediatrics107:690–692, 2001.

Feta

l Ped

iatr

Pat

hol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Cal

ifor

nia

Irvi

ne o

n 10

/31/

14Fo

r pe

rson

al u

se o

nly.

TNF-α AND IL-10 GENE POLYMORPHISMS VERSUS CARDIOIMMUNOLOGICAL RESPONSES IN SUDDEN INFANT DEATH

Documents

Transcript of TNF-α AND IL-10 GENE POLYMORPHISMS VERSUS CARDIOIMMUNOLOGICAL RESPONSES IN SUDDEN INFANT DEATH